Closed SGeeversAtVortech closed 1 month ago
In GitLab by @jvande42b on Jun 27, 2018, 05:55
@vreeken obviously I am not going to be spending much time developing anymore, but here is a change I squeezed in today. Feel free to take over this MR.
In GitLab by @jvande42b on Jun 27, 2018, 05:58
I am not sure this even counts as an API change, as I consider the old behavior a buggy attempt at what this does.
In GitLab by @vreeken on Jul 5, 2018, 21:37
added 3 commits
master
In GitLab by @baayen on Jul 10, 2018, 10:18
I'd suggest to update the description to suggest that this concerns simulation mode. I didn't understand the description at first (parameters unused? what?) until I read the code. :-)
In GitLab by @baayen on Jul 10, 2018, 10:22
Commented on src/rtctools/simulation/simulation_problem.py line 212
What if this parameter depends on another one, not specified in the model, which is set later on during this loop?
In GitLab by @jvande42b on Jul 10, 2018, 22:24
added 1 commit
In GitLab by @vreeken on Aug 28, 2019, 22:42
Closing as no progress for 1 year.
In GitLab by @vreeken on Aug 28, 2019, 22:42
closed
In GitLab by @jvande42b on Jun 27, 2018, 05:53
Before this commit, the parameters method was implemented by all the simulation mixins, but went unused. This resulted in confusing behaviour. Furthermore, the various mixins would set parameters in their own overwritten implementations of initialize().
With this commit, simulation child classes and mixins can override the parameters() method to set parameters instead of needing to implement and maintian their own set_var() logic. This also makes it much more transparent to the user what the values of the parameters are and which class takes precedence when setting parameters.